Erin Newburn, MS, PhD
Senior Manager, Field Applications Scientist

Ethnic Disparities in Precision Genomics

The premise of the right drug for the right patient at the right time is critical for indications from cardiovascular and infectious disease to the latest immuno-oncology strategies.  It’s imperative to comprehend which patient will derive benefit from a given treatment as therapeutic solutions expand.

Understanding which treatment will work for a disease subtype is clinically critical, but better recognition of whether the treatment will work for a particular individual is equally pressing.  Within this need comes the further necessity to determine genomic differences that may pertain to particular ethnic groups.  As results from whole exome and whole genome studies begin to flood the databases, the initial extension of genomic data beyond non-European populations has been seen.  Despite the significant benefits that continue to be gained from genomic sequencing, dedicated efforts are needed to ensure greater representation of minorities in these clinical trials and translational research efforts. 

Generalities in ethnic genomic trends have been suggested in cancer.  For instance, there is a higher incidence of triple negative breast cancer in African American women (Newman and Kaljee, 2017).  Likewise, men of African descent have a higher risk for prostate cancer (Nwaneri et al., 2016).  Additionally, Asians with lung cancer tend to have increased presence of EGFR mutations (Li et al., 2016).  This information can help tailor treatment strategy, adjust clinical trials, or make more precise screening recommendations.  However, a continued, conscious effort to include all ethnicities in our future genomic studies must be undertaken so that the benefits of these precision genomics can be realized by all of the world’s population.  Unfortunately, there have been several recent examples in the literature where ethnic disparities in cancer genomics can be noted. These results emphasize the call for diverse and additional populations in oncological genetic studies.

  • The Cancer Genome Atlas: Spratt et al., 2016 analyzed TCGA datasets from 10 different tumor types that included 5,729 patient samples. The majority of samples were white (77%), while only 12% were from African Americans, 3% Asian, and 3% Hispanic.  With such limited numbers from minorities, it would be difficult to identify specific mutations at low allele frequencies such as 10 or 5%.  A lack of statistical power is a common challenge in data generated from minority groups.
  • Clinical Tumor Profiling: Garofalo et al., 2016 showed that in clinical cancer tumor testing, in studies where the matched germline sample was not included, there was a higher percentage of false positive results in non-European patients.  Germline databases for certain ethnicities were shown to be limiting, yet utilizing molecular pathology review was helpful in alleviating this issue.
  • Immunotherapy: Data from the latest immunotherapy clinical trials has further revealed the extent of this limitation.  For example, Borghaei et al., 2015 in a trial comparing Nivolumab and Docetaxel in almost 600 Non-Small-Cell Lung Cancer (NSCLC) patients, only 3 percent of individuals were black and also only 3 percent were Asian. Likewise, in a larger Renal Cell Carcinoma study comparing Nivolumab and Everolimus (Motzer et al., 2015) in over 800 patients, just 1% (5 patients) were black.  Also, in a Phase III trial comparing Pembrolizumab and Ipilimumab in advanced melanoma patients, out of the 834 individuals approximately 6% were non-white (Robert et al., 2015).  These statistics may be unexpected to some and alarming at first glance, but awareness will help set the stage to address this issue.

For immunotherapy in particular, the struggle to find appropriate biomarkers and biomarker signatures has been evident.  Yet, it will be interesting to understand if the prospective markers like PD-L1, Tumor Mutational Burden, Neoantigen Load, or Tumor Infiltrating Lymphocytes for predicting checkpoint inhibitor responses have higher or lower correlation when analyzed within ethnicities? Some initial reports have shown evidence of this in a pan-cancer study examining PD-L1 expression in tumor infiltrating immune cells (Zhao et al., 2017).  Only in non-Asian cancer patients was the expression of PD-L1 in tumor infiltrating immune cells shown as an indicator of a favorable prognosis.

How do we address the issue at hand? The US Cancer Moonshot initiative has recognized the demand for diversity in Precision Medicine within oncology.  As this exciting effort aims to solve some of the greatest challenges in cancer, the group specified it would use this tremendous opportunity to address ethnic disparities.  Further, the collaborative nature promoted within the Cancer Moonshot to create a “cancer data ecosystem” for sharing and analyzing genomic data will also help to further research discovery in this regard.  The NIH Policy and Guidelines on the Inclusion of Women and Minorities as Subjects in Clinical Research is another effort created with a strong belief of the importance in studying diverse, global populations in genetic research.  The scientific community continues to evaluate this policy and propose further updates, such as more specific funding and recruitment strategies to address scientists’ concerns (Knerr et al., 2011).  Again, a concerted effort will help us to understand and work to eliminate these inadequacies. As many large-scale sequencing efforts are currently underway, we seek to unravel the genomics of cancer, but also to gain insight into those genomic traits that may be attributed to ethnic differences.